Policy communication and public compliance in environmental public opinion governance: a survey experiment based on multiple policy combinations and artificial intelligence applications

Purpose This study aims to examine how different policy communication approaches influence public compliance in environmental incident governance, specifically investigating: (1) the comparative effectiveness of expert versus artificial intelligence (AI) opinions in technology-oriented policy propaganda; (2) the relative impact of democratic versus technology-oriented approaches; and (3) the combined effects of different policy propaganda orientations and response modes. Design/methodology/approach This study used a two-experiment design examining policy communication’s influence on public compliance in environmental governance. Experiment 1 (n = 129) investigated expert versus AI opinions through a single-factor between-subjects design. Experiment 2 (n = 258) used a 2 × 3 factorial design exploring interactions between policy propaganda orientations (technology vs democratic) and response modes (control-oriented, consultative and nonresponse). Data collection was conducted via an online research platform with randomized participant assignment. Findings Results demonstrate that expert opinions are more effective than AI opinions in reducing public willingness to participate in mass incidents. Democratic-oriented policy propaganda proves more effective than technology-oriented approaches after rumor emergence. The combination of democratic-oriented propaganda with consultative response produces optimal outcomes in reducing mass incident participation willingness, while government nonresponse yields poorest results regardless of propaganda orientation. Originality/value This research contributes to risk communication theory by introducing the concept of “primacy bias,” extends policy combination theory through examining dynamic instrument interactions and enriches Science and Technology studies by exploring differentiated impacts of AI versus expert authority in environmental governance. The findings provide practical implications for environmental public opinion governance and policy communication strategies.